X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Ft-test.q;h=d5349dddf6f1a9b0431ef70c59266f9de0a757e5;hb=ef05645131b8163a4bb9839e609bb76d553b80e9;hp=0f4133a6c70577588637f88e7ac0ceb9cc776390;hpb=b9e28aa5614a079548c616bcf97aa804024ad647;p=pspp-builds.git

diff --git a/src/t-test.q b/src/t-test.q
index 0f4133a6..d5349ddd 100644
--- a/src/t-test.q
+++ b/src/t-test.q
@@ -27,7 +27,9 @@
 #include <math.h>
 #include "alloc.h"
 #include "str.h"
+#include "case.h"
 #include "command.h"
+#include "dictionary.h"
 #include "lexer.h"
 #include "error.h"
 #include "magic.h"
@@ -38,13 +40,15 @@
 #include "var.h"
 #include "vfm.h"
 #include "hash.h"
-#include "t-test.h"
+#include "group_proc.h"
+#include "casefile.h"
 #include "levene.h"
+/* (headers) */
 
 /* (specification)
    "T-TEST" (tts_):
      +groups=custom;
-     +testval=double;
+     testval=double;
      variables=varlist("PV_NO_SCRATCH | PV_NUMERIC");
      pairs=custom;
      +missing=miss:!analysis/listwise,
@@ -56,7 +60,7 @@
 /* (functions) */
 
 
-static struct cmd_t_test cmd;
+
 
 /* Function to use for testing for missing values */
 static is_missing_func value_is_missing;
@@ -64,12 +68,34 @@ static is_missing_func value_is_missing;
 /* Variable for the GROUPS subcommand, if given. */
 static struct variable *indep_var;
 
-/* GROUPS: Number of values specified by the user; the values
-   specified if any. */
+enum comparison
+  {
+    CMP_LE = -2,
+    CMP_EQ = 0,
+  };
 
-static int n_group_values;
-static union value groups_values[2];
-static enum comparison criteria[2];
+struct group_properties
+{
+  /* The comparison criterion */
+  enum comparison criterion;
+
+  /* The width of the independent variable */
+  int indep_width ;  
+
+  union {
+    /* The value of the independent variable at which groups are determined to 
+       belong to one group or the other */
+    double critical_value;
+    
+
+    /* The values of the independent variable for each group */
+    union value g_value[2];
+  } v ;
+
+};
+
+
+static struct group_properties gp ;
 
 
 
@@ -181,35 +207,48 @@ enum {
 };
 
 
-static int common_calc (struct ccase *, void *);
-static void common_precalc (void *);
-static void common_postcalc (void *);
+static int common_calc (const struct ccase *, void *);
+static void common_precalc (struct cmd_t_test *);
+static void common_postcalc (struct cmd_t_test *);
 
-static int one_sample_calc (struct ccase *, void *);
-static void one_sample_precalc (void *);
-static void one_sample_postcalc (void *);
+static int one_sample_calc (const struct ccase *, void *);
+static void one_sample_precalc (struct cmd_t_test *);
+static void one_sample_postcalc (struct cmd_t_test *);
 
-static int  paired_calc (struct ccase *, void *);
-static void paired_precalc (void *);
-static void paired_postcalc (void *);
+static int  paired_calc (const struct ccase *, void *);
+static void paired_precalc (struct cmd_t_test *);
+static void paired_postcalc (struct cmd_t_test *);
 
-static void group_precalc (void *);
-static int  group_calc (struct ccase *, void *);
-static void group_postcalc (void *);
+static void group_precalc (struct cmd_t_test *);
+static int  group_calc (const struct ccase *, struct cmd_t_test *);
+static void group_postcalc (struct cmd_t_test *);
 
 
 static int compare_var_name (const void *a_, const void *b_, void *v_ UNUSED);
 static unsigned hash_var_name (const void *a_, void *v_ UNUSED);
 
+static void calculate(const struct casefile *cf, void *_mode);
+
+static  int mode;
+
+static struct cmd_t_test cmd;
+
+static int bad_weight_warn;
+
+
+static int compare_group_binary(const struct group_statistics *a, 
+				const struct group_statistics *b, 
+				struct group_properties *p);
+
+
+static unsigned  hash_group_binary(const struct group_statistics *g, 
+				   struct group_properties *p);
+
 
 
 int
 cmd_t_test(void)
 {
-  int mode;
-
-  struct ssbox stat_summary_box;
-  struct trbox test_results_box;
 
   if ( !parse_t_test(&cmd) )
     return CMD_FAILURE;
@@ -291,36 +330,9 @@ cmd_t_test(void)
   else
     value_is_missing = is_missing;
 
-  procedure_with_splits (common_precalc, common_calc, common_postcalc, NULL);
-
-  switch(mode)
-    {
-    case T_1_SAMPLE:
-      procedure_with_splits (one_sample_precalc, one_sample_calc,
-                             one_sample_postcalc, NULL);
-      break;
-    case T_PAIRED:
-      procedure_with_splits (paired_precalc, paired_calc, paired_postcalc,
-                             NULL);
-      break;
-    case T_IND_SAMPLES:
-      procedure_with_splits(group_precalc,group_calc,group_postcalc, NULL);
-      levene(indep_var, cmd.n_variables, cmd.v_variables,
-	     (cmd.miss == TTS_LISTWISE)?LEV_LISTWISE:LEV_ANALYSIS ,
-	     value_is_missing);
-      break;
-    }
+  bad_weight_warn = 1;
 
-  ssbox_create(&stat_summary_box,&cmd,mode);
-  ssbox_populate(&stat_summary_box,&cmd);
-  ssbox_finalize(&stat_summary_box);
-
-  if ( mode == T_PAIRED) 
-      pscbox();
-
-  trbox_create(&test_results_box,&cmd,mode);
-  trbox_populate(&test_results_box,&cmd);
-  trbox_finalize(&test_results_box);
+  multipass_procedure_with_splits (calculate, &cmd);
 
   n_pairs=0;
   free(pairs);
@@ -328,11 +340,12 @@ cmd_t_test(void)
 
   if ( mode == T_IND_SAMPLES) 
     {
-      int i;
+      int v;
       /* Destroy any group statistics we created */
-      for (i= 0 ; i < cmd.n_variables ; ++i ) 
+      for (v = 0 ; v < cmd.n_variables ; ++v ) 
 	{
-	  free(cmd.v_variables[i]->p.t_t.gs);
+	  struct group_proc *grpp = group_proc_get (cmd.v_variables[v]);
+	  free(grpp->group_hash);
 	}
     }
     
@@ -342,6 +355,7 @@ cmd_t_test(void)
 static int
 tts_custom_groups (struct cmd_t_test *cmd UNUSED)
 {
+  int n_group_values=0;
 
   lex_match('=');
 
@@ -371,10 +385,13 @@ tts_custom_groups (struct cmd_t_test *cmd UNUSED)
     {
       if (indep_var->type == NUMERIC)
 	{
-	  groups_values[0].f = 1;
-	  groups_values[1].f = 2;
-	  criteria[0] = criteria[1] = CMP_EQ;
+	  gp.v.g_value[0].f = 1;
+	  gp.v.g_value[1].f = 2;
+
+	  gp.criterion = CMP_EQ;
+	  
 	  n_group_values = 2;
+
 	  return 1;
 	}
       else
@@ -385,27 +402,32 @@ tts_custom_groups (struct cmd_t_test *cmd UNUSED)
 	}
     }
 
-  if (!parse_value (&groups_values[0],indep_var->type))
+  if (!parse_value (&gp.v.g_value[0],indep_var->type))
       return 0;
 
   lex_match (',');
   if (lex_match (')'))
     {
-      criteria[0] =  CMP_LE;
-      criteria[1] =  CMP_GT;
-      groups_values[1] = groups_values[0];
+      gp.criterion = CMP_LE;
+      gp.v.critical_value = gp.v.g_value[0].f;
+
       n_group_values = 1;
       return 1;
     }
 
-  if (!parse_value (&groups_values[1],indep_var->type))
+  if (!parse_value (&gp.v.g_value[1],indep_var->type))
     return 0;
-  
+
   n_group_values = 2;
   if (!lex_force_match (')'))
     return 0;
 
-  criteria[0] = criteria[1] = CMP_EQ;
+  if ( n_group_values == 2 ) 
+    gp.criterion = CMP_EQ ;
+  else
+    gp.criterion = CMP_LE ;
+
+
   return 1;
 }
 
@@ -536,6 +558,7 @@ tts_custom_pairs (struct cmd_t_test *cmd UNUSED)
 
   n_pairs+=n_pairs_local;
 
+  free (vars);
   return 1;
 }
 
@@ -554,7 +577,7 @@ parse_value (union value * v, int type )
     {
       if (!lex_force_string ())
 	return 0;
-      strncpy (v->s, ds_value (&tokstr), ds_length (&tokstr));
+      strncpy (v->s, ds_c_str (&tokstr), ds_length (&tokstr));
     }
 
   lex_get ();
@@ -578,6 +601,7 @@ void ssbox_independent_samples_init(struct ssbox *this,
 void ssbox_paired_init(struct ssbox *this,
 			   struct cmd_t_test *cmd);
 
+
 /* Factory to create an ssbox */
 void 
 ssbox_create(struct ssbox *ssb, struct cmd_t_test *cmd, int mode)
@@ -599,6 +623,7 @@ ssbox_create(struct ssbox *ssb, struct cmd_t_test *cmd, int mode)
 }
 
 
+
 /* Despatcher for the populate method */
 void
 ssbox_populate(struct ssbox *ssb,struct cmd_t_test *cmd)
@@ -622,6 +647,8 @@ ssbox_base_finalize(struct ssbox *ssb)
   tab_submit(ssb->t);
 }
 
+
+
 /* Initialize a ssbox struct */
 void 
 ssbox_base_init(struct ssbox *this, int cols,int rows)
@@ -651,7 +678,7 @@ ssbox_one_sample_init(struct ssbox *this,
 
   ssbox_base_init(this, hsize,vsize);
   tab_title (this->t, 0, _("One-Sample Statistics"));
-  tab_vline(this->t, TAL_2, 1,0,vsize);
+  tab_vline(this->t, TAL_2, 1,0,vsize - 1);
   tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE, _("N"));
   tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _("Mean"));
   tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
@@ -673,7 +700,7 @@ ssbox_independent_samples_init(struct ssbox *this,
 
   ssbox_base_init(this, hsize,vsize);
   tab_title (this->t, 0, _("Group Statistics"));
-  tab_vline(this->t,0,1,0,vsize);
+  tab_vline(this->t,0,1,0,vsize - 1);
   tab_text (this->t, 1, 0, TAB_CENTER | TAT_TITLE, indep_var->name);
   tab_text (this->t, 2, 0, TAB_CENTER | TAT_TITLE, _("N"));
   tab_text (this->t, 3, 0, TAB_CENTER | TAT_TITLE, _("Mean"));
@@ -691,31 +718,41 @@ ssbox_independent_samples_populate(struct ssbox *ssb,
 
   char *val_lab0=0;
   char *val_lab1=0;
+  double indep_value[2];
 
   char prefix[2][3]={"",""};
 
   if ( indep_var->type == NUMERIC ) 
     {
-      val_lab0 = val_labs_find( indep_var->val_labs,groups_values[0]); 
-      val_lab1 = val_labs_find( indep_var->val_labs,groups_values[1]);
+      val_lab0 = val_labs_find( indep_var->val_labs,gp.v.g_value[0]); 
+      val_lab1 = val_labs_find( indep_var->val_labs,gp.v.g_value[1]);
     }
   else
     {
-      val_lab0 = groups_values[0].s;
-      val_lab1 = groups_values[1].s;
+      val_lab0 = gp.v.g_value[0].s;
+      val_lab1 = gp.v.g_value[1].s;
     }
 
-  if (n_group_values == 1) 
+  if (gp.criterion == CMP_LE ) 
     {
       strcpy(prefix[0],"< ");
       strcpy(prefix[1],">=");
+      indep_value[0] = gp.v.critical_value;
+      indep_value[1] = gp.v.critical_value;
+    }
+  else
+    {
+      indep_value[0] = gp.v.g_value[0].f;
+      indep_value[1] = gp.v.g_value[1].f;
     }
 
   assert(ssb->t);
 
   for (i=0; i < cmd->n_variables; ++i)
     {
-      int g;
+      struct variable *var = cmd->v_variables[i];
+      struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
+      int count=0;
 
       tab_text (ssb->t, 0, i*2+1, TAB_LEFT, cmd->v_variables[i]->name);
 
@@ -723,26 +760,50 @@ ssbox_independent_samples_populate(struct ssbox *ssb,
 	tab_text (ssb->t, 1, i*2+1, TAB_LEFT | TAT_PRINTF, 
 		  "%s%s", prefix[0], val_lab0);
       else
-	tab_text (ssb->t, 1, i*2+1, TAB_LEFT | TAT_PRINTF, 
-		  "%s%g", prefix[0], groups_values[0].f); 
+	  tab_text (ssb->t, 1, i*2+1, TAB_LEFT | TAT_PRINTF, 
+		    "%s%g", prefix[0], indep_value[0]);
 
 
       if (val_lab1)
 	tab_text (ssb->t, 1, i*2+1+1, TAB_LEFT | TAT_PRINTF, 
 		  "%s%s", prefix[1], val_lab1);
       else
-	tab_text (ssb->t, 1, i*2+1+1, TAB_LEFT | TAT_PRINTF, 
-		  "%s%g", prefix[1], groups_values[1].f); 
+	  tab_text (ssb->t, 1, i*2+1+1, TAB_LEFT | TAT_PRINTF, 
+		    "%s%g", prefix[1], indep_value[1]);
+
 
       /* Fill in the group statistics */
-      for ( g=0; g < 2 ; ++g ) 
+      for ( count = 0 ; count < 2 ; ++count ) 
 	{
-	  struct group_statistics *gs = &cmd->v_variables[i]->p.t_t.gs[g];
+	  union value search_val;
+
+	  struct group_statistics *gs;
 
-	  tab_float(ssb->t, 2 ,i*2+g+1, TAB_RIGHT, gs->n, 2, 0);
-	  tab_float(ssb->t, 3 ,i*2+g+1, TAB_RIGHT, gs->mean, 8, 2);
-	  tab_float(ssb->t, 4 ,i*2+g+1, TAB_RIGHT, gs->std_dev, 8, 3);
-	  tab_float(ssb->t, 5 ,i*2+g+1, TAB_RIGHT, gs->se_mean, 8, 3);
+	  if ( gp.criterion == CMP_LE ) 
+	    {
+	      if ( count == 0 ) 
+		{
+		  /*  less than ( < )  case */
+		  search_val.f = gp.v.critical_value - 1.0;
+		}
+	      else
+		{
+		  /* >= case  */
+		  search_val.f = gp.v.critical_value + 1.0;
+		}
+	    }
+	  else
+	    {
+	      search_val = gp.v.g_value[count];
+	    }
+
+	  gs = hsh_find(grp_hash, (void *) &search_val);
+	  assert(gs);
+
+	  tab_float(ssb->t, 2 ,i*2+count+1, TAB_RIGHT, gs->n, 2, 0);
+	  tab_float(ssb->t, 3 ,i*2+count+1, TAB_RIGHT, gs->mean, 8, 2);
+	  tab_float(ssb->t, 4 ,i*2+count+1, TAB_RIGHT, gs->std_dev, 8, 3);
+	  tab_float(ssb->t, 5 ,i*2+count+1, TAB_RIGHT, gs->se_mean, 8, 3);
 	}
     }
 }
@@ -790,7 +851,7 @@ ssbox_paired_populate(struct ssbox *ssb,struct cmd_t_test *cmd UNUSED)
 	{
 	  struct group_statistics *gs;
 
-	  gs=&pairs[i].v[j]->p.t_t.ugs;
+	  gs = &group_proc_get (pairs[i].v[j])->ugs;
 
 	  /* Titles */
 
@@ -816,8 +877,7 @@ ssbox_one_sample_populate(struct ssbox *ssb, struct cmd_t_test *cmd)
 
   for (i=0; i < cmd->n_variables; ++i)
     {
-      struct group_statistics *gs;
-      gs= &cmd->v_variables[i]->p.t_t.ugs;
+      struct group_statistics *gs = &group_proc_get (cmd->v_variables[i])->ugs;
 
       tab_text (ssb->t, 0, i+1, TAB_LEFT, cmd->v_variables[i]->name);
       tab_float (ssb->t,1, i+1, TAB_RIGHT, gs->n, 2, 0);
@@ -925,8 +985,8 @@ trbox_independent_samples_init(struct trbox *self,
   tab_text(self->t,10,2, TAB_CENTER | TAT_TITLE,_("Upper"));
 
   tab_joint_text(self->t, 9, 1, 10, 1, TAB_CENTER | TAT_PRINTF, 
-		 _("%d%% Confidence Interval of the Difference"),
-		 (int)round(cmd->criteria*100.0));
+		 _("%g%% Confidence Interval of the Difference"),
+		 cmd->criteria*100.0);
 
 }
 
@@ -951,21 +1011,43 @@ trbox_independent_samples_populate(struct trbox *self,
       double std_err_diff;
       double mean_diff;
 
-      struct group_statistics *gs0 = &cmd->v_variables[i]->p.t_t.gs[0];
-      struct group_statistics *gs1 = &cmd->v_variables[i]->p.t_t.gs[1];
+      struct variable *var = cmd->v_variables[i];
+      struct group_proc *grp_data = group_proc_get (var);
+
+      struct hsh_table *grp_hash = grp_data->group_hash;
+
+      struct group_statistics *gs0 ;
+      struct group_statistics *gs1 ;
+	  
+      union value search_val;
+	  
+      if ( gp.criterion == CMP_LE ) 
+	search_val.f = gp.v.critical_value - 1.0;
+      else
+	search_val = gp.v.g_value[0];
+
+      gs0 = hsh_find(grp_hash, (void *) &search_val);
+      assert(gs0);
+
+      if ( gp.criterion == CMP_LE ) 
+	search_val.f = gp.v.critical_value + 1.0;
+      else
+	search_val = gp.v.g_value[1];
+
+      gs1 = hsh_find(grp_hash, (void *) &search_val);
+      assert(gs1);
+
 	  
       tab_text (self->t, 0, i*2+3, TAB_LEFT, cmd->v_variables[i]->name);
 
       tab_text (self->t, 1, i*2+3, TAB_LEFT, _("Equal variances assumed"));
 
 
-      tab_float(self->t, 2, i*2+3, TAB_CENTER, 
-		cmd->v_variables[i]->p.t_t.levene, 8,3);
-
+      tab_float(self->t, 2, i*2+3, TAB_CENTER, grp_data->levene, 8,3);
 
       /* Now work out the significance of the Levene test */
-      df1 = 1; df2 = cmd->v_variables[i]->p.t_t.ugs.n - 2;
-      q = gsl_cdf_fdist_Q(cmd->v_variables[i]->p.t_t.levene, df1, df2);
+      df1 = 1; df2 = grp_data->ugs.n - 2;
+      q = gsl_cdf_fdist_Q(grp_data->levene, df1, df2);
 
       tab_float(self->t, 3, i*2+3, TAB_CENTER, q, 8,3 );
 
@@ -999,7 +1081,6 @@ trbox_independent_samples_populate(struct trbox *self,
       q = (1 - cmd->criteria)/2.0;  /* 2-tailed test */
 
       t = gsl_cdf_tdist_Qinv(q,df);
-
       tab_float(self->t, 9, i*2+3, TAB_RIGHT, 
 		mean_diff - t * std_err_diff, 8, 3); 
 
@@ -1071,7 +1152,7 @@ trbox_paired_init(struct trbox *self,
   trbox_base_init(self,n_pairs,hsize);
   tab_title (self->t, 0, _("Paired Samples Test"));
   tab_hline(self->t,TAL_1,2,6,1);
-  tab_vline(self->t,TAL_2,2,0,vsize);
+  tab_vline(self->t,TAL_2,2,0,vsize - 1);
   tab_joint_text(self->t,2,0,6,0,TAB_CENTER,_("Paired Differences"));
   tab_box(self->t,-1,-1,-1,TAL_1, 2,1,6,vsize-1);
   tab_box(self->t,-1,-1,-1,TAL_1, 6,0,hsize-1,vsize-1);
@@ -1079,8 +1160,8 @@ trbox_paired_init(struct trbox *self,
   tab_vline(self->t,TAL_0,6,0,1);
 
   tab_joint_text(self->t, 5, 1, 6, 1, TAB_CENTER | TAT_PRINTF, 
-		 _("%d%% Confidence Interval of the Difference"),
-		 (int)round(cmd->criteria*100.0));
+		 _("%g%% Confidence Interval of the Difference"),
+		 cmd->criteria*100.0);
 
   tab_text (self->t, 2, 2, TAB_CENTER | TAT_TITLE, _("Mean"));
   tab_text (self->t, 3, 2, TAB_CENTER | TAT_TITLE, _("Std. Deviation"));
@@ -1165,17 +1246,17 @@ trbox_one_sample_init(struct trbox *self, struct cmd_t_test *cmd )
   trbox_base_init(self, cmd->n_variables,hsize);
   tab_title (self->t, 0, _("One-Sample Test"));
   tab_hline(self->t, TAL_1, 1, hsize - 1, 1);
-  tab_vline(self->t, TAL_2, 1, 0, vsize);
+  tab_vline(self->t, TAL_2, 1, 0, vsize - 1);
 
   tab_joint_text(self->t, 1, 0, hsize-1,0, TAB_CENTER | TAT_PRINTF, 
-		 _("Test Value = %f"),cmd->n_testval);
+		 _("Test Value = %f"), cmd->n_testval[0]);
 
   tab_box(self->t, -1, -1, -1, TAL_1, 1,1,hsize-1,vsize-1);
 
 
   tab_joint_text(self->t,5,1,6,1,TAB_CENTER  | TAT_PRINTF, 
-		 _("%d%% Confidence Interval of the Difference"),
-		 (int)round(cmd->criteria*100.0));
+		 _("%g%% Confidence Interval of the Difference"),
+		 cmd->criteria*100.0);
 
   tab_vline(self->t,TAL_0,6,1,1);
   tab_hline(self->t,TAL_1,5,6,2);
@@ -1202,13 +1283,12 @@ trbox_one_sample_populate(struct trbox *trb, struct cmd_t_test *cmd)
       double t;
       double p,q;
       double df;
-      struct group_statistics *gs;
-      gs= &cmd->v_variables[i]->p.t_t.ugs;
+      struct group_statistics *gs = &group_proc_get (cmd->v_variables[i])->ugs;
 
 
       tab_text (trb->t, 0, i+3, TAB_LEFT, cmd->v_variables[i]->name);
 
-      t = (gs->mean - cmd->n_testval ) * sqrt(gs->n) / gs->std_dev ;
+      t = (gs->mean - cmd->n_testval[0] ) * sqrt(gs->n) / gs->std_dev ;
 
       tab_float (trb->t, 1, i+3, TAB_RIGHT, t, 8,3);
 
@@ -1320,24 +1400,26 @@ pscbox(void)
 
 
 
+
 /* Calculation Implementation */
 
 /* Per case calculations common to all variants of the T test */
 static int 
-common_calc (struct ccase *c, void *aux UNUSED)
+common_calc (const struct ccase *c, void *_cmd)
 {
   int i;
+  struct cmd_t_test *cmd = (struct cmd_t_test *)_cmd;  
 
-  double weight = dict_get_case_weight(default_dict,c);
+  double weight = dict_get_case_weight(default_dict,c,&bad_weight_warn);
 
 
   /* Skip the entire case if /MISSING=LISTWISE is set */
-  if ( cmd.miss == TTS_LISTWISE ) 
+  if ( cmd->miss == TTS_LISTWISE ) 
     {
-      for(i=0; i< cmd.n_variables ; ++i) 
+      for(i=0; i< cmd->n_variables ; ++i) 
 	{
-	  struct variable *v = cmd.v_variables[i];
-	  union value *val = &c->data[v->fv];
+	  struct variable *v = cmd->v_variables[i];
+	  const union value *val = case_data (c, v->fv);
 
 	  if (value_is_missing(val,v) )
 	    {
@@ -1347,9 +1429,9 @@ common_calc (struct ccase *c, void *aux UNUSED)
     }
 
   /* Listwise has to be implicit if the independent variable is missing ?? */
-  if ( cmd.sbc_groups )
+  if ( cmd->sbc_groups )
     {
-      union value *gv = &c->data[indep_var->fv];
+      const union value *gv = case_data (c, indep_var->fv);
       if ( value_is_missing(gv,indep_var) )
 	{
 	  return 0;
@@ -1357,13 +1439,13 @@ common_calc (struct ccase *c, void *aux UNUSED)
     }
 
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      struct variable *v = cmd.v_variables[i];
-      union value *val = &c->data[v->fv];
+      struct variable *v = cmd->v_variables[i];
+      const union value *val = case_data (c, v->fv);
 
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
 
       if (! value_is_missing(val,v) )
 	{
@@ -1377,14 +1459,14 @@ common_calc (struct ccase *c, void *aux UNUSED)
 
 /* Pre calculations common to all variants of the T test */
 static void 
-common_precalc (void *aux UNUSED)
+common_precalc ( struct cmd_t_test *cmd )
 {
   int i=0;
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
       
       gs->sum=0;
       gs->n=0;
@@ -1395,14 +1477,15 @@ common_precalc (void *aux UNUSED)
 
 /* Post calculations common to all variants of the T test */
 void 
-common_postcalc (void *aux UNUSED)
+common_postcalc (  struct cmd_t_test *cmd )
 {
   int i=0;
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
       
       gs->mean=gs->sum / gs->n;
       gs->s_std_dev= sqrt(
@@ -1421,19 +1504,21 @@ common_postcalc (void *aux UNUSED)
 
 /* Per case calculations for one sample t test  */
 static int 
-one_sample_calc (struct ccase *c, void *aux UNUSED)
+one_sample_calc (const struct ccase *c, void *cmd_)
 {
   int i;
+  struct cmd_t_test *cmd = (struct cmd_t_test *)cmd_;
+
 
-  double weight = dict_get_case_weight(default_dict,c);
+  double weight = dict_get_case_weight(default_dict,c,&bad_weight_warn);
 
   /* Skip the entire case if /MISSING=LISTWISE is set */
-  if ( cmd.miss == TTS_LISTWISE ) 
+  if ( cmd->miss == TTS_LISTWISE ) 
     {
-      for(i=0; i< cmd.n_variables ; ++i) 
+      for(i=0; i< cmd->n_variables ; ++i) 
 	{
-	  struct variable *v = cmd.v_variables[i];
-	  union value *val = &c->data[v->fv];
+	  struct variable *v = cmd->v_variables[i];
+	  const union value *val = case_data (c, v->fv);
 
 	  if (value_is_missing(val,v) )
 	    {
@@ -1442,16 +1527,16 @@ one_sample_calc (struct ccase *c, void *aux UNUSED)
 	}
     }
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      struct variable *v = cmd.v_variables[i];
-      union value *val = &c->data[v->fv];
+      struct variable *v = cmd->v_variables[i];
+      const union value *val = case_data (c, v->fv);
 
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
       
       if ( ! value_is_missing(val,v))
-	gs->sum_diff += weight * (val->f - cmd.n_testval);
+	gs->sum_diff += weight * (val->f - cmd->n_testval[0]);
     }
 
   return 0;
@@ -1459,14 +1544,14 @@ one_sample_calc (struct ccase *c, void *aux UNUSED)
 
 /* Pre calculations for one sample t test */
 static void 
-one_sample_precalc (void *aux UNUSED)
+one_sample_precalc ( struct cmd_t_test *cmd )
 {
-  int i=0;
-  
-  for(i=0; i< cmd.n_variables ; ++i) 
+  int i=0; 
+ 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
       
       gs->sum_diff=0;
     }
@@ -1474,16 +1559,15 @@ one_sample_precalc (void *aux UNUSED)
 
 /* Post calculations for one sample t test */
 static void 
-one_sample_postcalc (void *aux UNUSED)
+one_sample_postcalc (struct cmd_t_test *cmd)
 {
   int i=0;
   
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
       struct group_statistics *gs;
-      gs= &cmd.v_variables[i]->p.t_t.ugs;
+      gs= &group_proc_get (cmd->v_variables[i])->ugs;
 
-      
       gs->mean_diff = gs->sum_diff / gs->n ;
     }
 }
@@ -1510,7 +1594,7 @@ hash_var_name (const void *a_, void *v_ UNUSED)
 
 
 static void 
-paired_precalc (void *aux UNUSED)
+paired_precalc (struct cmd_t_test *cmd UNUSED)
 {
   int i;
 
@@ -1529,23 +1613,25 @@ paired_precalc (void *aux UNUSED)
 
 
 static int  
-paired_calc (struct ccase *c, void *aux UNUSED)
+paired_calc (const struct ccase *c, void *cmd_)
 {
   int i;
 
-  double weight = dict_get_case_weight(default_dict,c);
+  struct cmd_t_test *cmd  = (struct cmd_t_test *) cmd_;
+
+  double weight = dict_get_case_weight(default_dict,c,&bad_weight_warn);
 
   /* Skip the entire case if /MISSING=LISTWISE is set , 
    AND one member of a pair is missing */
-  if ( cmd.miss == TTS_LISTWISE ) 
+  if ( cmd->miss == TTS_LISTWISE ) 
     {
       for(i=0; i < n_pairs ; ++i )
       	{
 	  struct variable *v0 = pairs[i].v[0];
 	  struct variable *v1 = pairs[i].v[1];
 
-	  union value *val0 = &c->data[v0->fv];
-	  union value *val1 = &c->data[v1->fv];
+	  const union value *val0 = case_data (c, v0->fv);
+	  const union value *val1 = case_data (c, v1->fv);
 	  
 	  if ( value_is_missing(val0,v0) ||
 	       value_is_missing(val1,v1) )
@@ -1560,8 +1646,8 @@ paired_calc (struct ccase *c, void *aux UNUSED)
       struct variable *v0 = pairs[i].v[0];
       struct variable *v1 = pairs[i].v[1];
 
-      union value *val0 = &c->data[v0->fv];
-      union value *val1 = &c->data[v1->fv];
+      const union value *val0 = case_data (c, v0->fv);
+      const union value *val1 = case_data (c, v1->fv);
 
       if ( ( !value_is_missing(val0,v0) && !value_is_missing(val1,v1) ) )
       {
@@ -1583,7 +1669,7 @@ paired_calc (struct ccase *c, void *aux UNUSED)
 }
 
 static void 
-paired_postcalc (void *aux UNUSED)
+paired_postcalc (struct cmd_t_test *cmd UNUSED)
 {
   int i;
 
@@ -1621,94 +1707,76 @@ paired_postcalc (void *aux UNUSED)
     }
 }
 
-/* Return the group # corresponding to the 
-   independent variable with the value val 
-*/
-static int
-get_group(const union value *val, struct variable *indep)
-{
-  int i; 
-
-  for (i = 0; i < 2  ; ++i )
-    {
-      const int cmp = compare_values(val,&groups_values[i],indep->width) ;
-      switch ( criteria[i])
-	{
-	case CMP_EQ: 
-	  if ( 0 == cmp )   return i;
-	  break;	   
-	case CMP_LT:	   
-	  if ( 0 >  cmp )  return i;
-	  break;	   
-	case CMP_LE:	   
-	  if ( cmp <= 0 )   return i;
-	  break;
-	case CMP_GT:
-	  if ( cmp > 0 ) return i;
-	  break;
-	case CMP_GE:
-	  if ( cmp >= 0 ) return i;
-	  break;
-	default:
-	  assert(0);
-	};
-    }
-
-  /* No groups matched */
-  return -1;
-}
-
-
 static void 
-group_precalc (void *aux UNUSED)
+group_precalc (struct cmd_t_test *cmd )
 {
   int i;
   int j;
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
-      struct t_test_proc *ttpr = &cmd.v_variables[i]->p.t_t;
+      struct group_proc *ttpr = group_proc_get (cmd->v_variables[i]);
 
       /* There's always 2 groups for a T - TEST */
       ttpr->n_groups = 2;
-      ttpr->gs = xmalloc(sizeof(struct group_statistics) * 2) ;
+
+      gp.indep_width = indep_var->width;
+      
+      ttpr->group_hash = hsh_create(2, 
+				    (hsh_compare_func *) compare_group_binary,
+				    (hsh_hash_func *) hash_group_binary,
+				    (hsh_free_func *) free_group,
+				    (void *) &gp );
 
       for (j=0 ; j < 2 ; ++j)
 	{
-	  ttpr->gs[j].sum = 0;
-	  ttpr->gs[j].n = 0;
-	  ttpr->gs[j].ssq = 0;
+
+	  struct group_statistics *gs = (struct group_statistics *) 
+	    xmalloc (sizeof(struct group_statistics));
+
+	  gs->sum = 0;
+	  gs->n = 0;
+	  gs->ssq = 0;
 	
-	  if ( n_group_values == 2 ) 
-	    ttpr->gs[j].id = groups_values[j];
+	  if ( gp.criterion == CMP_EQ ) 
+	    {
+	      gs->id = gp.v.g_value[j];
+	    }
 	  else
-	    ttpr->gs[j].id = groups_values[0];
-	  ttpr->gs[j].criterion = criteria[j];
+	    {
+	      if ( j == 0 ) 
+		gs->id.f = gp.v.critical_value - 1.0 ;
+	      else
+		gs->id.f = gp.v.critical_value + 1.0 ;
+	    }
+	  
+	  hsh_insert ( ttpr->group_hash, (void *) gs );
+
 	}
     }
 
 }
 
 static int  
-group_calc (struct ccase *c, void *aux UNUSED)
+group_calc (const struct ccase *c, struct cmd_t_test *cmd)
 {
   int i;
-  int g;
-  union value *gv = &c->data[indep_var->fv];
 
-  double weight = dict_get_case_weight(default_dict,c);
+  const union value *gv = case_data (c, indep_var->fv);
+
+  const double weight = dict_get_case_weight(default_dict,c,&bad_weight_warn);
 
   if ( value_is_missing(gv,indep_var) )
     {
       return 0;
     }
 
-  if ( cmd.miss == TTS_LISTWISE ) 
+  if ( cmd->miss == TTS_LISTWISE ) 
     {
-      for(i=0; i< cmd.n_variables ; ++i) 
+      for(i=0; i< cmd->n_variables ; ++i) 
 	{
-	  struct variable *v = cmd.v_variables[i];
-	  union value *val = &c->data[v->fv];
+	  struct variable *v = cmd->v_variables[i];
+	  const union value *val = case_data (c, v->fv);
 
 	  if (value_is_missing(val,v) )
 	    {
@@ -1717,23 +1785,21 @@ group_calc (struct ccase *c, void *aux UNUSED)
 	}
     }
 
+  gv = case_data (c, indep_var->fv);
 
-  gv = &c->data[indep_var->fv];
-
-  g = get_group(gv,indep_var);
-
-  /* If the independent variable doesn't match either of the values 
-     for this case then move on to the next case */
-  if (g == -1 ) 
-    return 0;
-
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
-      struct variable *var = cmd.v_variables[i];
+      struct variable *var = cmd->v_variables[i];
+      const union value *val = case_data (c, var->fv);
+      struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
+      struct group_statistics *gs;
 
-      struct group_statistics *gs = &var->p.t_t.gs[g];
+      gs = hsh_find(grp_hash, (void *) gv);
 
-      union value *val=&c->data[var->fv];
+      /* If the independent variable doesn't match either of the values 
+         for this case then move on to the next case */
+      if ( ! gs ) 
+      	return 0;
 
       if ( !value_is_missing(val,var) )
 	{
@@ -1748,31 +1814,175 @@ group_calc (struct ccase *c, void *aux UNUSED)
 
 
 static void 
-group_postcalc (void *aux UNUSED)
+group_postcalc ( struct cmd_t_test *cmd )
 {
   int i;
-  int j;
 
-  for(i=0; i< cmd.n_variables ; ++i) 
+  for(i=0; i< cmd->n_variables ; ++i) 
     {
-      for (j=0 ; j < 2 ; ++j)
-	{
-	  struct group_statistics *gs;
-	  gs=&cmd.v_variables[i]->p.t_t.gs[j];
+      struct variable *var = cmd->v_variables[i];
+      struct hsh_table *grp_hash = group_proc_get (var)->group_hash;
+      struct hsh_iterator g;
+      struct group_statistics *gs;
+      int count=0;
 
+      for (gs =  hsh_first (grp_hash,&g); 
+	   gs != 0; 
+	   gs = hsh_next(grp_hash,&g))
+	{
 	  gs->mean = gs->sum / gs->n;
 	  
 	  gs->s_std_dev= sqrt(
-			 ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
-			 ) ;
+			      ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
+			      ) ;
 
 	  gs->std_dev= sqrt(
-			 gs->n/(gs->n-1) *
-			 ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
-			 ) ;
+			    gs->n/(gs->n-1) *
+			    ( (gs->ssq / gs->n ) - gs->mean * gs->mean )
+			    ) ;
 	  
 	  gs->se_mean = gs->std_dev / sqrt(gs->n);
+	  count ++;
+	}
+      assert(count == 2);
+    }
+}
+
+
+
+static void 
+calculate(const struct casefile *cf, void *cmd_)
+{
+  struct ssbox stat_summary_box;
+  struct trbox test_results_box;
+
+  struct casereader *r;
+  struct ccase c;
+
+  struct cmd_t_test *cmd = (struct cmd_t_test *) cmd_;
+
+  common_precalc(cmd);
+  for(r = casefile_get_reader (cf);
+      casereader_read (r, &c) ;
+      case_destroy (&c)) 
+    {
+      common_calc(&c,cmd);
+    }
+  casereader_destroy (r);
+  common_postcalc(cmd);
+
+  switch(mode)
+    {
+    case T_1_SAMPLE:
+      one_sample_precalc(cmd);
+      for(r = casefile_get_reader (cf);
+	  casereader_read (r, &c) ;
+          case_destroy (&c)) 
+	{
+	  one_sample_calc(&c,cmd);
 	}
+      casereader_destroy (r);
+      one_sample_postcalc(cmd);
+
+      break;
+    case T_PAIRED:
+      paired_precalc(cmd);
+      for(r = casefile_get_reader (cf);
+	  casereader_read (r, &c) ;
+          case_destroy (&c)) 
+	{
+	  paired_calc(&c,cmd);
+	}
+      casereader_destroy (r);
+      paired_postcalc(cmd);
+
+      break;
+    case T_IND_SAMPLES:
+
+      group_precalc(cmd);
+      for(r = casefile_get_reader (cf);
+	  casereader_read (r, &c) ;
+          case_destroy (&c)) 
+	{
+	  group_calc(&c,cmd);
+	}
+      casereader_destroy (r);
+      group_postcalc(cmd);
+
+      levene(cf, indep_var, cmd->n_variables, cmd->v_variables,
+	     (cmd->miss == TTS_LISTWISE)?LEV_LISTWISE:LEV_ANALYSIS ,
+	     value_is_missing);
+      break;
     }
+
+  ssbox_create(&stat_summary_box,cmd,mode);
+  ssbox_populate(&stat_summary_box,cmd);
+  ssbox_finalize(&stat_summary_box);
+
+  if ( mode == T_PAIRED) 
+      pscbox();
+
+  trbox_create(&test_results_box,cmd,mode);
+  trbox_populate(&test_results_box,cmd);
+  trbox_finalize(&test_results_box);
+
 }
 
+
+/* Return -1 if the id of a is less than b; +1 if greater than and 
+   0 if equal */
+static int 
+compare_group_binary(const struct group_statistics *a, 
+		     const struct group_statistics *b, 
+		     struct group_properties *p)
+{
+  
+  short flag_a;
+  short flag_b;
+
+  assert(p->indep_width == 0 ) ;
+
+  if ( p->criterion == CMP_LE ) 
+    {
+      flag_a = ( a->id.f < p->v.critical_value ) ;
+      flag_b = ( b->id.f < p->v.critical_value ) ;
+    }
+  else
+    {
+      flag_a = ( a->id.f == p->v.critical_value ) ;
+      flag_b = ( b->id.f == p->v.critical_value ) ;
+    }
+     
+
+  if ( flag_a == flag_b) 
+    return 0 ;
+  
+  return ( flag_a < flag_b);
+}
+
+static unsigned 
+hash_group_binary(const struct group_statistics *g, struct group_properties *p)
+{
+  short flag = -1;
+
+  assert(p->indep_width == 0 ) ;
+
+      /* FIXME: should compare union values */    
+  if ( p->criterion == CMP_LE ) 
+    {
+      flag = ( g->id.f < p->v.critical_value ) ; 
+    }
+  else if ( p->criterion == CMP_EQ) 
+    {
+      if ( g->id.f ==  p->v.g_value[0].f ) 
+	flag = 0 ;
+      else if ( g->id.f == p->v.g_value[1].f ) 
+	flag = 1;
+      else
+	flag = 2;
+    }
+  else
+    assert(0);
+
+  return flag;
+}